|
[DPRG] Balancing robot difficulties
Subject: [DPRG] Balancing robot difficulties
From: dpa
dpa at io.isem.smu.edu
Date: Tue Jul 3 14:07:39 CDT 2007
Hi Luke,
First of all, congratulations! You've done the hard
part: getting the robot to balance.
Looks to me like you are right about backlash in the
gears. That would explain the twitchy behavior. I've
seen the same thing on some of my balancer experiments.
Two-wheel balance, particularly stationary, puts very
tight requirements on the acceptable backlash of the
gear train, because the motors are constantly and
rapidly changing direction, unlike most robot
applications. Even when driving at a steady
velocity, the balancing algorithm is continuously
loading and unloading the gears to keep the robot
in balance.
Turns out that this is not a trivial problem to solve.
Planetary gears have very little backlash, so that's one
possible solution. The Segway uses "harmonic drives"
to get rid of the backlash problem with 2 wheel balance,
not quite sure how that works but it looks like a variation
on planetary gears, makes for nice flat "pancake" motors,
and probably costs arm+leg.
I think you may have to go to some more expensive motors
and gearheads, unfortunately.
High friction in the drive train can also cause the kind of
overshoot visible in the video. But its more likely a
problem with slop in the gear train, as you surmised.
Keep up the great work!
best regards,
dpa
More information about the DPRG mailing list
|
